[
Lists Home |
Date Index |
Thread Index
]
Simon St.Laurent wrote:
> jcowan@reutershealth.com (John Cowan) writes:
> >I think no one would object to this solution with the exception of
> >IANA, which would have the job of actually maintaining the registry.
> >IANA is overworked and underfunded.
>
> I'm not convinced that the problem is large enough to require IANA
> activity. MIME Media Type registrations already go through IANA, and
> have associated RFCs which define fragment identifiers. In some sense,
> the IETF already has this problem well-covered for most cases.
Yes, the way I think about it is that anyone registering a media type of the
form:
application/foo+xml
might define a fragement identifier syntax which uses the suggested syntax
as in
http://www.rddl.org/fragment-syntax which is dated but might be updated
given the recent work in XPointer.
Ultimately the syntax used for fragment identifiers is media type dependent,
XPointer provides a solution for application/xml (and an extension
mechanism), so ultimately if you are willing to go to the trouble to define
your own media type, you can have whatever syntax you like. I am not sure
that XPointer *prevents* you for doing that but if it does, then *you* (and
your coauthors) might change that by modifying
http://www.ietf.org/rfc/rfc3023.txt appropriately. Perhaps RFC 3023 is due
for an updating (concerning the fragment identifier syntax and semantics)
now that XPointer has made more progress.
Jonathan
http://www.jonathanborden-md.com
http://www.erieneurosurgery.com
http://www.openhealth.org
|
